CN108021691A - Answer lookup method, customer service robot and computer-readable recording medium - Google Patents
Answer lookup method, customer service robot and computer-readable recording medium Download PDFInfo
- Publication number
- CN108021691A CN108021691A CN201711363315.2A CN201711363315A CN108021691A CN 108021691 A CN108021691 A CN 108021691A CN 201711363315 A CN201711363315 A CN 201711363315A CN 108021691 A CN108021691 A CN 108021691A
- Authority
- CN
- China
- Prior art keywords
- answer
- question
- wait
- target
- knowledge base
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/243—Natural language query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Finance (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of answer lookup method, customer service robot and computer-readable recording medium, the method comprising the steps of:When getting when answering a question, to it is described wait to answer a question pre-process, obtain pretreated described waiting to answer a question;Judge whether to find in knowledge base according to preset rules and described wait corresponding target problem of answering a question with pretreated;If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained, and export the target answer.The present invention, which realizes, to be got when answering a question, first treat to answer a question and handled, then corresponding target problem is being searched in knowledge base according to preset rules, to determine corresponding answer, add answer in knowledge base and search mode, improve the accuracy rate that customer problem is answered by customer service robot.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of answer lookup method, customer service robot and calculating
Machine readable storage medium storing program for executing.
Background technology
Existing customer service machine is manually as flow:When client machine people receives the problem of user proposes, client computer
The problem of device people will be received is matched with the vocabulary pattern set in its knowledge base.If found in knowledge base with being somebody's turn to do
The corresponding vocabulary pattern of problem, then return to user by the corresponding answer of vocabulary pattern;If do not found in knowledge base
Vocabulary pattern corresponding with the problem, then return to the acquiescence answer pre-set in knowledge base to user.In customer service robot
Knowledge base composition it is complicated, conflict is easily produced between various vocabulary patterns, and to search mode single for answer, so as to cause client
During answer corresponding with customer problem is searched from knowledge base, the accuracy rate for searching answer is low for robot.
The content of the invention
It is a primary object of the present invention to provide a kind of answer lookup method, customer service robot and computer-readable storage
Medium, it is intended to solve the low technical problem of accuracy rate that existing customer service robot answers a question.
To achieve the above object, the present invention provides a kind of answer lookup method, and the answer lookup method includes step:
When getting when answering a question, to it is described wait to answer a question pre-process, obtain pretreated described treat
Answer a question;
According to preset rules judge whether to find in knowledge base with it is pretreated it is described wait to answer a question it is corresponding
Target problem;
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And export the target answer.
Preferably, it is described judge whether to find in knowledge base according to preset rules described wait to answer with pretreated
After the step of problem corresponding target problem, further include:
If not finding the target problem in the knowledge base, judge whether to find with advance locating in storehouse is chatted
Corresponding target problem of answering a question is waited described in after reason;
If finding the target problem in the chat storehouse, export target corresponding with the target problem and answer
Case;
If not finding the target problem in the chat storehouse, the acquiescence answer to prestore is obtained, and described in output
Give tacit consent to answer.
Preferably, if described find the target problem in the knowledge base, obtain the target problem and correspond to
Target answer, and the step of exporting the target answer include:
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And obtain related question corresponding with the target problem;
Export the target answer and the related question.
Preferably, it is described when getting when answering a question, to it is described wait to answer a question pre-process, pre-processed
After the step of to be answered a question described in afterwards, further include:
By it is pretreated it is described wait to answer a question stored with the target answer into the knowledge base.
Preferably, it is described when getting when answering a question, to it is described wait to answer a question pre-process, pre-processed
The step of to be answered a question described in afterwards, includes:
When getting when answering a question, to it is described wait to answer a question clean, described after being cleaned waits to answer
Problem;
Completion is carried out to waiting described in after cleaning to answer a question, obtain after completion described waits to answer a question;
Classify to waiting to answer a question described in after completion, the generic to be answered a question is determined, to obtain
Wait to answer a question described in pretreated.
Preferably, it is described when getting when answering a question, to it is described wait to answer a question clean, after being cleaned
The step of described to be answered a question, includes:
When getting when answering a question, to it is described wait to answer a question carry out word segmentation processing, obtain described waiting to answer a question
Corresponding vocabulary;
Calculate the first similarity between the vocabulary and default vocabulary;
If first similarity is more than predetermined threshold value, the corresponding vocabulary of first similarity is deleted, it is clear to obtain
Wait to answer a question described in after washing.
Preferably, described after described pair of cleaning, which waits to answer a question, carries out completion, and obtain after completion described treats that answer is asked
The step of topic, includes:
Word segmentation processing is carried out to waiting described in after cleaning to answer a question, obtains described waiting corresponding vocabulary of answering a question;
The vocabulary and preset group word rule are contrasted, obtain comparing result;
Wait to answer a question described in after being cleaned according to the comparing result completion, to treat that answer is asked described in obtaining after completion
Topic.
Preferably, it is described judge whether to find in knowledge base according to preset rules described wait to answer with pretreated
The step of problem corresponding target problem, includes:
Determine in the knowledge base with it is pretreated it is described wait to answer a question belong to same category of same problems, and count
The second similarity answered a question between any same problems is waited described in calculation;
If there is second similarity for being more than default similarity, by the corresponding same problems of maximum second similarity
As target problem;
If there is no the second similarity more than the default similarity, it is determined that wait to answer with described in the knowledge base
The corresponding relevant issues of problem;
The third phase for waiting to answer a question between any relevant issues described in calculating is like degree;
If there is the third phase more than the default similarity like spending, maximum third phase is seemingly spent into corresponding relevant issues
As target problem;
If there is no the third phase more than the default similarity like spending, default knowledge mapping is obtained;
Based on the knowledge mapping, the problem of traveling through in the knowledge base of answering a question is waited by described;
If obtaining traversing result, the problem of traversing result is corresponded to, is as the target problem.
In addition, to achieve the above object, the present invention also provides a kind of customer service robot, the customer service robot includes storage
Device, processor and the answer search program that can be run on the memory and on the processor is stored in, the answer is looked into
The step of answer lookup method as described above is realized when looking for program to be performed by the processor.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer-readable recording medium
Answer search program is stored with storage medium, the answer search program realizes answer as described above when being executed by processor
The step of lookup method.
The present invention by when getting when answering a question, to it is described wait to answer a question pre-process, pre-processed
Wait to answer a question described in afterwards;Judge whether to find in knowledge base according to preset rules and described wait to answer with pretreated
The corresponding target problem of problem;If finding the target problem in the knowledge base, obtain the target problem and correspond to
Target answer, and export the target answer.Realize and getting when answering a question, first treat to answer a question and located
Reason, is then searching corresponding target problem according to preset rules in knowledge base, to determine corresponding answer, adds knowledge
Mode is searched in answer in storehouse, improves the accuracy rate that customer problem is answered by customer service robot.
Brief description of the drawings
Fig. 1 is the system structure diagram for the hardware running environment that the embodiment of the present invention is related to;
Fig. 2 is the flow diagram of answer lookup method first embodiment of the present invention;
Fig. 3 be in the embodiment of the present invention when getting when answering a question, to it is described wait to answer a question pre-process, obtain
To a kind of pretreated flow diagram to be answered a question;
Fig. 4 is the flow diagram of answer lookup method second embodiment of the present invention;
If Fig. 5 is that the target problem is found in the knowledge base in the embodiment of the present invention, the target is obtained
The corresponding target answer of problem, and export a kind of flow diagram of the target answer.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The solution of the embodiment of the present invention is mainly:When getting when answering a question, to it is described wait to answer a question into
Row pretreatment, obtains pretreated described waiting to answer a question;According to preset rules judge whether to find in knowledge base with
Corresponding target problem of answering a question is waited described in pretreated;If finding the target problem in the knowledge base,
The corresponding target answer of the target problem is obtained, and exports the target answer.Answered a question with solving customer service robot
The problem of accuracy rate is low.
As shown in Figure 1, the system structure diagram for the hardware running environment that Fig. 1, which is the embodiment of the present invention, to be related to.
Customer service of embodiment of the present invention robot can be PC or smart mobile phone, tablet computer, pocket computer etc.
Terminal device.
As shown in Figure 1, the customer service robot can include:Processor 1001, such as CPU, network interface 1004, Yong Hujie
Mouth 1003, memory 1005, communication bus 1002.Wherein, the connection that communication bus 1002 is used for realization between these components is led to
Letter.User interface 1003 can include display screen (Display), input unit such as keyboard (Keyboard), and optional user connects
Mouth 1003 can also include standard wireline interface and wireless interface.Network interface 1004 can optionally include the wired of standard
Interface, wave point (such as WI-FI interfaces).Memory 1005 can be high-speed RAM memory or the memory of stabilization
(non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of aforementioned processor
1001 storage device.
Alternatively, customer service robot can also include camera, RF (Radio Frequency, radio frequency) circuit, sensing
Device, voicefrequency circuit, WiFi module etc..
It will be understood by those skilled in the art that the restriction of the customer service robot architecture shown in Fig. 1 not structure paired terminal,
It can include than illustrating more or fewer components, either combine some components or different components arrangement.
As shown in Figure 1, including operating system and it can be answered as in a kind of memory 1005 of computer-readable storage medium
Case search program.Wherein, operating system is the program of management and control customer service robot hardware and software resource, supports answer to look into
Look for the operation of program and other softwares and/or program.
In the customer service robot shown in Fig. 1, network interface 1004 is mainly used for accessing network;User interface 1003 is main
Wait to answer a question for obtaining.And processor 1001 can be used for calling the answer search program stored in memory 1005, and
Perform following operation:
When getting when answering a question, to it is described wait to answer a question pre-process, obtain pretreated described treat
Answer a question;
According to preset rules judge whether to find in knowledge base with it is pretreated it is described wait to answer a question it is corresponding
Target problem;
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And export the target answer.
Further, it is described judge whether to find in knowledge base according to preset rules described treated back with pretreated
After question and answer inscribes the step of corresponding target problem, processor 1001 can be also used for calling the answer stored in memory 1005
Search program, and perform following steps:
If not finding the target problem in the knowledge base, judge whether to find with advance locating in storehouse is chatted
Corresponding target problem of answering a question is waited described in after reason;
If finding the target problem in the chat storehouse, export target corresponding with the target problem and answer
Case;
If not finding the target problem in the chat storehouse, the acquiescence answer to prestore is obtained, and described in output
Give tacit consent to answer.
Further, if described find the target problem in the knowledge base, the target problem pair is obtained
The target answer answered, and the step of exporting the target answer include:
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And obtain related question corresponding with the target problem;
Export the target answer and the related question.
Further, it is described when getting when answering a question, to it is described wait to answer a question pre-process, obtain pre- place
After the step of to be answered a question described in after reason, processor 1001 can be also used for calling the answer stored in memory 1005
Search program, and perform following steps:
By it is pretreated it is described wait to answer a question stored with the target answer into the knowledge base.
Further, it is described when getting when answering a question, to it is described wait to answer a question pre-process, obtain pre- place
The step of to be answered a question described in after reason, includes:
When getting when answering a question, to it is described wait to answer a question clean, described after being cleaned waits to answer
Problem;
Completion is carried out to waiting described in after cleaning to answer a question, obtain after completion described waits to answer a question;
Classify to waiting to answer a question described in after completion, the generic to be answered a question is determined, to obtain
Wait to answer a question described in pretreated.
Further, it is described when getting when answering a question, to it is described wait to answer a question clean, after obtaining cleaning
It is described to be answered a question the step of include:
When getting when answering a question, to it is described wait to answer a question carry out word segmentation processing, obtain described waiting to answer a question
Corresponding vocabulary;
Calculate the first similarity between the vocabulary and default vocabulary;
If first similarity is more than predetermined threshold value, the corresponding vocabulary of first similarity is deleted, it is clear to obtain
Wait to answer a question described in after washing.
Further, described after described pair of cleaning, which waits to answer a question, carries out completion, and obtain after completion described waits to answer
The step of problem, includes:
Word segmentation processing is carried out to waiting described in after cleaning to answer a question, obtains described waiting corresponding vocabulary of answering a question;
The vocabulary and preset group word rule are contrasted, obtain comparing result;
Wait to answer a question described in after being cleaned according to the comparing result completion, to treat that answer is asked described in obtaining after completion
Topic.
Further, it is described judge whether to find in knowledge base according to preset rules described treated back with pretreated
The step of question and answer topic corresponding target problem, includes:
Determine in the knowledge base with it is pretreated it is described wait to answer a question belong to same category of same problems, and count
The second similarity answered a question between any same problems is waited described in calculation;
If there is second similarity for being more than default similarity, by the corresponding same problems of maximum second similarity
As target problem;
If there is no the second similarity more than the default similarity, it is determined that wait to answer with described in the knowledge base
The corresponding relevant issues of problem;
The third phase for waiting to answer a question between any relevant issues described in calculating is like degree;
If there is the third phase more than the default similarity like spending, maximum third phase is seemingly spent into corresponding relevant issues
As target problem;
If there is no the third phase more than the default similarity like spending, default knowledge mapping is obtained;
Based on the knowledge mapping, the problem of traveling through in the knowledge base of answering a question is waited by described;
If obtaining traversing result, the problem of traversing result is corresponded to, is as the target problem.
Based on above-mentioned hardware configuration, each embodiment of proposition answer lookup method.
With reference to Fig. 2, Fig. 2 is the flow diagram of answer lookup method first embodiment of the present invention.
In the present embodiment, there is provided the embodiment of answer lookup method is, it is necessary to illustrate, although showing in flow charts
Go out logical order, but in some cases, can be with the steps shown or described are performed in an order that is different from the one herein.
The answer lookup method includes:
Step S10, when getting when answering a question, to it is described wait to answer a question pre-process, after obtaining pretreatment
Described wait to answer a question.
When customer service robot gets input by user when answering a question, it is pre- that customer service robot treats progress of answering a question
Processing, to obtain pretreated waiting to answer a question.Wherein, user can be manually entered in client machine people's display interface and treat
Answer a question, user can also be inputted by way of voice and wait to answer a question.When customer service robot gets speech form
, can waiting speech form to answer a question and be converted to written form and wait to answer a question when answering a question.
Further, include with reference to Fig. 3, step S10:
Step S11, when getting when answering a question, to it is described wait to answer a question clean, the institute after being cleaned
State and wait to answer a question.
Step S12, completion is carried out to waiting described in after cleaning to answer a question, and obtain after completion described waits to answer a question.
Step S13, classifies to waiting to answer a question described in after completion, determines the affiliated class to be answered a question
Not, to obtain pretreated described waiting to answer a question.
Customer service robot treats the detailed process answered a question and pre-processed:Wait to answer when customer service robot is got
During problem, treat to answer a question and cleaned, wait to answer a question after being cleaned.Treated after customer service robot obtains cleaning
After answering a question, customer service robot obtains waiting to answer a question after completion, and to completion to waiting to answer and carry out completion after cleaning
Afterwards wait to answer a question is classified, and to determine generic to be answered a question, obtains pretreated waiting to answer a question.
Classification to be answered a question is stored in advance in knowledge base for customer service robot.When customer service robot detects volume
When collecting the edit instruction of classification, new classification can be increased according to edit instruction, delete existing classification, or the existing class of modification
Not.Such as classification may be configured as " loan class ", " interest class " and " financing class ";It may be alternatively provided as " explanation class ", " reason class " etc..
In the progress assorting process of waiting to answer a question after to completion, it can determine to treat back according to the vocabulary waited in answering a question
The generic of question and answer topic.In knowledge base, each classification has oneself corresponding keyword.Containing in waiting to answer a question should
During keyword, that is, think that this waits to answer a question and belong to keyword place classification.As contained keyword " particulate in " loan class "
Borrow ", when wait answer a question for " particulate of how refunding loan " when, due to waiting to contain " particulate loan " in answering a question, can determine that
This, which waits to answer a question, belongs to " loan class ".
Further, step S11 includes:
Step a, when getting when answering a question, waits progress word segmentation processing of answering a question to described, obtains described treat back
Question and answer inscribes corresponding vocabulary.
Step b, calculates the first similarity between the vocabulary and default vocabulary;
Step c, if first similarity is more than predetermined threshold value, deletes the corresponding vocabulary of first similarity, with
Wait to answer a question described in after being cleaned.
When customer service robot is got when answering a question, customer service robot, which treats to answer a question, carries out word segmentation processing, obtains
To waiting corresponding vocabulary of answering a question.Specifically, customer service robot can be treated by participle instrument of increasing income answers a question point
Word processing.When customer service robot obtain wait answer a question corresponding vocabulary when, customer service robot calculates equivalent to be answered a question
The similarity between default vocabulary in remittance and knowledge base, is denoted as the first similarity.Specifically, customer service robot can pass through cosine
Similarity measure goes out the first similarity between the default vocabulary in wait to answer a question corresponding vocabulary and knowledge base.If treating, answer is asked
The first similarity between the corresponding vocabulary of topic and default vocabulary is more than predetermined threshold value, and similarity correspondence is then deleted by customer service robot
Vocabulary, to wait to answer a question after being cleaned;If it is first similar between corresponding vocabulary and default vocabulary to wait to answer a question
Degree is less than or equal to predetermined threshold value, and customer service robot then retains the corresponding vocabulary of the first similarity.It is it should be noted that pre-
If vocabulary thinks the vocabulary nonsensical to determining answer to prestore into knowledge base, by customer service robot, such as " thanks ",
The vocabulary such as " manual service ", " not having ", " hello ", " turn artificial " and "no".Predetermined threshold value can be set according to specific needs,
60% is such as may be configured as, or is arranged to 76% etc..
As when wait to answer a question for " it is what that particulate, which is borrowed, be thanks " when, treat after answering a question and carrying out word segmentation processing,
Obtained vocabulary is " particulate loan ", "Yes", " what ", " thing " and " thanks ".Wait to answer a question as " particulate is borrowed after cleaning
What ".
Further, before progress word segmentation processing of answering a question is treated, customer service robot, which first detects, to be treated for customer service robot
With the presence or absence of letter, numeral and/or punctuation mark in answering a question.If customer service robot detects wait to answer a question in there are word
The letter waited in answering a question, numeral and/or punctuation mark are then deleted by female, numeral and/or punctuation mark, customer service robot;Visitor
Treat answer if taking robot and letter, numeral and/or punctuation mark, customer service robot being not detected by waiting to answer a question and ask
Topic carries out word segmentation processing.
Further, step S12 includes:
Step d, word segmentation processing is carried out to waiting described in after cleaning to answer a question, and obtains described waiting corresponding word of answering a question
Converge.
Step e, the vocabulary and preset group word rule are contrasted, obtain comparing result.
Step f, waits to answer a question described in after being cleaned according to the comparing result completion, to treat described in obtaining after completion
Answer a question.
Detailed process to be answered a question after customer service robot completion cleaning is:After customer service robot obtains cleaning
When answering a question, customer service robot, which treats to answer a question, carries out word segmentation processing, obtains waiting corresponding vocabulary of answering a question, will treat
Corresponding vocabulary of answering a question is contrasted with preset group word rule, obtains comparing result.Customer service robot is according to comparing result
Wait to answer a question after completion cleaning, to obtain waiting to answer a question after completion.
It should be noted that preset group word rule is the vocabulary group word rule pre-set in knowledge base.Such as " going back " word
With " money " word group word, " refund " can connect title of various loan products etc., or the vocabulary such as connection " date " below.Such as when clear
After washing wait answer a question for " how to go back particulate borrow " when, wait to answer a question as " particulate of how refunding loan " after completion.
In embodiments of the present invention, treat to answer a question during completion and carry out word segmentation processing with being treated in cleaning process
The method for progress word segmentation processing of answering a question is consistent, and details are not described herein.It is understood that in the present embodiment, cleaning
During can not treat answer a question carry out word segmentation processing, and directly use cleaning process in perform delete operation after gained word
Converge as the vocabulary during completion.
Whether step S20, judge to find in knowledge base and described treat that answer is asked with pretreated according to preset rules
Inscribe corresponding target problem.
When customer service robot obtains pretreated when answering a question, customer service robot judges knowing according to preset rules
Know whether to find in storehouse and wait corresponding target problem of answering a question with pretreated.Wherein, preset rules include but unlimited
In problem matching, weight sequencing and collection of illustrative plates reasoning.
Further, step S20 includes:
Step g, determine in the knowledge base with it is pretreated it is described wait to answer a question belong to same category of and similar ask
Topic, and the second similarity answered a question between any same problems is waited described in calculating.
Step h, it is if there is second similarity for being more than default similarity, maximum second similarity is corresponding same
Class problem is as target problem.
The matched detailed process of problem is:When customer service robot obtains pretreated when answering a question, customer service machine
People by sorting algorithm determine in knowledge base with it is pretreated wait to answer a question belong to same category of same problems, and calculate
The second similarity for waiting to answer a question between any same problems, judges in the second similarity obtained by calculating with the presence or absence of big
In the second similarity of default similarity, if there is the second similarity for being more than default similarity, it is determined that exist in knowledge base
Target problem, at this time, using the corresponding same problems of the second similarity maximum in knowledge base as target problem.Need what is illustrated
It is that default similarity can be arranged as required to, such as could be provided as 50%, 70% or 85%.Sorting algorithm include but
It is not limited to K-NN (K Nearest Neighbours) sorting algorithm, SVM (Support Vector Machine, supporting vector
Learning machine) algorithm and logistic regression sorting algorithm.
Step i, if do not exist more than the default similarity the second similarity, it is determined that in the knowledge base with it is described
Wait corresponding relevant issues of answering a question.
Step j, the third phase for waiting described in calculating to answer a question between any relevant issues is like spending.
Step k, if there is the third phase more than the default similarity like spending, seemingly spends corresponding phase by maximum third phase
Pass problem is as target problem.
If not there is no the second similarity for being more than default similarity, by waiting the keyword in answering a question in knowledge base
Middle search contains the key word problem.It should be noted that it is to be answered in knowledge base with waiting containing the key word problem
The corresponding relevant issues of problem.The similarity for waiting to answer a question between any relevant issues is calculated, third phase is denoted as like spending, sentences
Disconnected whether there is seemingly is spent more than the third phase for presetting similarity., will most if there is the third phase for being more than default similarity like spending
Big third phase seemingly spends corresponding relevant issues as target problem.
Further, after relevant issues are determined, TF-IDF (term frequency-inverse can be passed through
Document frequency) algorithm calculates pretreated shared weight of waiting to answer a question in each relevant issues.
It is understood that in other embodiments, customer service robot can also be calculated by other algorithms similar to TF-IDF
Pretreated shared weight in each relevant issues of knowledge base of waiting to answer a question.Then selected in the weight obtained by calculating
The relevant issues of default quantity are selected, the similarity for waiting to answer a question between selected relevant issues is calculated, is denoted as third phase
Like degree.
Step l, if there is no the third phase more than the default similarity like spending, obtains default knowledge mapping.
Step m, based on the knowledge mapping, the problem of traveling through in the knowledge base of answering a question is waited by described.
Step n, if obtaining traversing result, the problem of traversing result is corresponded to, is as the target problem.
If there is no the third phase for being more than default similarity like spending, the knowledge mapping prestored in knowledge base is obtained,
Target problem is searched in knowledge base by knowledge mapping.Target problem detailed process is searched in knowledge base by knowledge mapping
For:Customer service robot uses default knowledge mapping, by waiting the problem of traveling through in knowledge base of answering a question, judges whether to obtain
Traversing result.If obtain traversing result, it is determined that the problem of traversing result corresponds to, the problem of traversing result is corresponded to is as target
Problem.
Prestored it should be noted that knowledge mapping is customer service robot, the reasoning for problem.Specifically, lead to
Pretreated triple to be answered a question is crossed, entity information corresponding with pending problem is searched in knowledge base, to push away
Manage out target problem to be answered a question.Wherein, triple is subject, predicate and object.
As when wait to answer a question as " what relation A and C is ", if " A and B are found in knowledge base by knowledge mapping
It is any relation " and the problem of " B with C be what relation ", customer service robot then determines " what relation A and B is " and " B is with C
What relation " is target problem." if what relation A is with B " corresponding answer is " A generates B ", " what relation B and C is "
Corresponding answer is " B generates C ", it follows that the corresponding target answer of target problem can be that " A generates B, and B is generated
C”。
Step S30, if finding the target problem in the knowledge base, obtains the corresponding mesh of the target problem
Answer is marked, and exports the target answer.
Step S40, if not finding the target problem in the knowledge base, obtains the acquiescence answer to prestore, and
Export the acquiescence answer.
When customer service robot found in knowledge base with it is pretreated wait answer a question corresponding target problem when, visitor
Take robot and target answer corresponding with the target problem is obtained in knowledge base, and the target answer is exported to it and shows boundary
In face, so that user checks.It should be noted that in knowledge base, the corresponding answer associated storage of each problem.Cause
This, when determining some problem in knowledge base, you can determine its corresponding answer.
When customer service robot do not searched in knowledge base with it is pretreated wait answer a question corresponding target problem when, visitor
Take robot and obtain the acquiescence answer of its storage that prestores, and export the acquiescence answer.Wherein, the content for giving tacit consent to answer can be according to tool
Body is needed and set, and such as may be configured as " sorry not find satisfactory answer, we understand continuous learning ".
The present embodiment by when getting when answering a question, to it is described wait to answer a question pre-process, obtain pre- place
Wait to answer a question described in after reason;Judge whether to find in knowledge base according to preset rules and described treated back with pretreated
Question and answer inscribes corresponding target problem;If finding the target problem in the knowledge base, the target problem pair is obtained
The target answer answered, and export the target answer.Realize and getting when answering a question, first treat progress of answering a question
Processing, is then searching corresponding target problem according to preset rules in knowledge base, to determine corresponding answer, adds and knows
Know answer in storehouse and search mode, improve the accuracy rate that customer problem is answered by customer service robot.
Further, answer lookup method second embodiment of the present invention is proposed.
Difference lies in reference with the answer lookup method first embodiment for the answer lookup method second embodiment
Fig. 4, answer lookup method further include:
Step S50, if not finding the target problem in the knowledge base, judges whether to search in storehouse is chatted
Described corresponding target problem of answering a question is waited to pretreated.
Step S60, if finding the target problem in the chat storehouse, output is corresponding with the target problem
Target answer.
Step S70, if not finding the target problem in the chat storehouse, obtains the acquiescence answer to prestore, and
Export the acquiescence answer.
When customer service robot do not found in knowledge base with it is pretreated wait answer a question corresponding target problem when,
Customer service robot judges whether to find in storehouse is chatted waits corresponding target problem of answering a question with pretreated.Wherein,
The problem of chatting in storehouse is the everyday problem for the user Chang Wen that customer service robot prestores.Specifically, which can be by visitor
Robot is taken to acquire in the data of user and customer service robot interaction by data mining.
When customer service robot found in chatting storehouse with it is pretreated wait answer a question corresponding target problem after, visitor
Take robot and corresponding target answer is determined according to target problem, and export the target answer, so that user checks.Need to illustrate
, customer service robot judges whether that the process that target problem is found in storehouse is chatted is searched with judging whether in knowledge base
The process of target problem is similar, and details are not described herein.
If client machine people does not find in storehouse is chatted waits corresponding target problem of answering a question, visitor with pretreated
Take robot and then obtain the acquiescence answer prestored, and export the acquiescence answer so that user checks.
The present embodiment by when customer service robot does not find target answer in knowledge base, go chat storehouse in search with
Target problem, only when not finding target problem in chatting storehouse, just output acquiescence answer, improves customer service robot and returns
Answer the success rate of customer problem.
Further, answer lookup method 3rd embodiment of the present invention is proposed.
Difference lies in reference with the answer lookup method first embodiment for the answer lookup method 3rd embodiment
Fig. 5, step S30 include:
Step S31, if finding the target problem in the knowledge base, obtains the corresponding mesh of the target problem
Answer is marked, and obtains related question corresponding with the target problem.
Step S32, exports the target answer and the related question.
If customer service robot finds target problem in knowledge base, customer service robot obtains the corresponding mesh of target problem
Mark answer, and obtain corresponding with target problem related question, export acquired in target answer and related question, for
Family is checked.Specifically, customer service robot can determine related question according to the keyword in target problem, wherein, the related question
For it is related to target problem the problem of.If " what be particulate borrow " target problem be, then can determine that keyword be " particulate loan ", logical
" particulate loan " confirmable related question is crossed as " interest that particulate is borrowed is how many ", " how applying for that particulate is borrowed " etc..
Further, customer service robot can obtain the related question of default quantity when obtaining related question, such as obtain three
Bar related question, or obtain five related questions.
The present embodiment exports target and asks by after target problem is determined, obtaining the corresponding related question of target problem
Topic and related question, so that user checks.User is realized during oneself corresponding answer of input problem is obtained, can be in visitor
The display interface for taking robot directly selects the problem of the problem of being inputted with it is associated, is asked in order to which user quickly obtains association
The answer of topic, improves the intelligent of customer service robot.
Further, answer lookup method further includes:
Step p, by it is pretreated it is described wait to answer a question stored with the target answer into the knowledge base.
When customer service robot obtain it is pretreated when answering a question, customer service robot will wait to answer a question with it is corresponding
Target answer is stored into knowledge base, the problem of to extend automatically in knowledge base and answer corresponding with the problem.
In addition, the embodiment of the present invention also proposes a kind of computer-readable recording medium, the computer-readable recording medium
On be stored with answer search program, the answer search program realizes following steps when being executed by processor:
When getting when answering a question, to it is described wait to answer a question pre-process, obtain pretreated described treat
Answer a question;
According to preset rules judge whether to find in knowledge base with it is pretreated it is described wait to answer a question it is corresponding
Target problem;
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And export the target answer.
Further, it is described judge whether to find in knowledge base according to preset rules described treated back with pretreated
After question and answer inscribes the step of corresponding target problem, the answer search program realizes following steps when being executed by processor:
If not finding the target problem in the knowledge base, judge whether to find with advance locating in storehouse is chatted
Corresponding target problem of answering a question is waited described in after reason;
If finding the target problem in the chat storehouse, export target corresponding with the target problem and answer
Case;
If not finding the target problem in the chat storehouse, the acquiescence answer to prestore is obtained, and described in output
Give tacit consent to answer.
Further, if described find the target problem in the knowledge base, the target problem pair is obtained
The target answer answered, and the step of exporting the target answer include:
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained,
And obtain related question corresponding with the target problem;
Export the target answer and the related question.
Further, it is described when getting when answering a question, to it is described wait to answer a question pre-process, obtain pre- place
After the step of to be answered a question described in after reason, the answer search program realizes following steps when being executed by processor:
By it is pretreated it is described wait to answer a question stored with the target answer into the knowledge base.
Further, it is described when getting when answering a question, to it is described wait to answer a question pre-process, obtain pre- place
The step of to be answered a question described in after reason, includes:
When getting when answering a question, to it is described wait to answer a question clean, described after being cleaned waits to answer
Problem;
Completion is carried out to waiting described in after cleaning to answer a question, obtain after completion described waits to answer a question;
Classify to waiting to answer a question described in after completion, the generic to be answered a question is determined, to obtain
Wait to answer a question described in pretreated.
Further, it is described when getting when answering a question, to it is described wait to answer a question clean, after obtaining cleaning
It is described to be answered a question the step of include:
When getting when answering a question, to it is described wait to answer a question carry out word segmentation processing, obtain described waiting to answer a question
Corresponding vocabulary;
Calculate the first similarity between the vocabulary and default vocabulary;
If first similarity is more than predetermined threshold value, the corresponding vocabulary of first similarity is deleted, it is clear to obtain
Wait to answer a question described in after washing.
Further, described after described pair of cleaning, which waits to answer a question, carries out completion, and obtain after completion described waits to answer
The step of problem, includes:
Word segmentation processing is carried out to waiting described in after cleaning to answer a question, obtains described waiting corresponding vocabulary of answering a question;
The vocabulary and preset group word rule are contrasted, obtain comparing result;
Wait to answer a question described in after being cleaned according to the comparing result completion, to treat that answer is asked described in obtaining after completion
Topic.
Further, it is described judge whether to find in knowledge base according to preset rules described treated back with pretreated
The step of question and answer topic corresponding target problem, includes:
Determine in the knowledge base with it is pretreated it is described wait to answer a question belong to same category of same problems, and count
The second similarity answered a question between any same problems is waited described in calculation;
If there is second similarity for being more than default similarity, by the corresponding same problems of maximum second similarity
As target problem;
If there is no the second similarity more than the default similarity, it is determined that wait to answer with described in the knowledge base
The corresponding relevant issues of problem;
The third phase for waiting to answer a question between any relevant issues described in calculating is like degree;
If there is the third phase more than the default similarity like spending, maximum third phase is seemingly spent into corresponding relevant issues
As target problem;
If there is no the third phase more than the default similarity like spending, default knowledge mapping is obtained;
Based on the knowledge mapping, the problem of traveling through in the knowledge base of answering a question is waited by described;
If obtaining traversing result, the problem of traversing result is corresponded to, is as target problem computer of the present invention
Readable storage medium storing program for executing embodiment and above-mentioned each embodiment of answer lookup method are essentially identical, and details are not described herein.
It should be noted that herein, term " comprising ", "comprising" or its any other variant are intended to non-row
His property includes, so that process, method, article or device including a series of elements not only include those key elements, and
And other elements that are not explicitly listed are further included, or further include as this process, method, article or device institute inherently
Key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including this
Also there are other identical element in the process of key element, method, article or device.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on such understanding, technical scheme substantially in other words does the prior art
Going out the part of contribution can be embodied in the form of software product, which is stored in a storage medium
In (such as ROM/RAM, magnetic disc, CD), including some instructions are used so that a station terminal equipment (can be mobile phone, computer, takes
Be engaged in device, air conditioner, or network equipment etc.) perform method described in each embodiment of the present invention.
It these are only the preferred embodiment of the present invention, be not intended to limit the scope of the invention, it is every to utilize this hair
The equivalent structure or equivalent flow shift that bright specification and accompanying drawing content are made, is directly or indirectly used in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of answer lookup method, it is characterised in that the answer lookup method comprises the following steps:
When getting when answering a question, to it is described wait to answer a question pre-process, obtain pretreated described waiting to answer
Problem;
Judge whether to find in knowledge base according to preset rules and described wait corresponding target of answering a question with pretreated
Problem;
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained, and it is defeated
Go out the target answer.
2. answer lookup method as claimed in claim 1, it is characterised in that described to be judged according to preset rules in knowledge base
Whether find with it is pretreated it is described wait to answer a question corresponding target problem the step of after, further include:
If not finding the target problem in the knowledge base, judge whether after finding and pre-processing in chatting storehouse
Described wait corresponding target problem of answering a question;
If finding the target problem in the chat storehouse, target answer corresponding with the target problem is exported;
If not finding the target problem in the chat storehouse, the acquiescence answer to prestore is obtained, and export the acquiescence
Answer.
3. answer lookup method as claimed in claim 1, it is characterised in that if it is described found in the knowledge base it is described
Target problem, then obtain the corresponding target answer of the target problem, and the step of exporting the target answer includes:
If finding the target problem in the knowledge base, the corresponding target answer of the target problem is obtained, and
Obtain related question corresponding with the target problem;
Export the target answer and the related question.
4. answer lookup method as claimed in claim 1, it is characterised in that it is described when getting when answering a question, to institute
State to wait to answer a question and pre-processed, after obtaining the step of pretreated described to be answered a question, further included:
By it is pretreated it is described wait to answer a question stored with the target answer into the knowledge base.
5. answer lookup method as claimed in claim 1, it is characterised in that it is described when getting when answering a question, to institute
State to wait to answer a question and pre-processed, obtaining the step of pretreated described to be answered a question includes:
When getting when answering a question, to it is described wait to answer a question clean, described after being cleaned waits to answer a question;
Completion is carried out to waiting described in after cleaning to answer a question, obtain after completion described waits to answer a question;
Classify to waiting to answer a question described in after completion, the generic to be answered a question is determined, to obtain pre- place
Wait to answer a question described in after reason.
6. answer lookup method as claimed in claim 5, it is characterised in that it is described when getting when answering a question, to institute
State to wait to answer a question and cleaned, the step of described to be answered a question after being cleaned includes:
When getting when answering a question, wait progress word segmentation processing of answering a question to described, obtain the correspondence to be answered a question
Vocabulary;
Calculate the first similarity between the vocabulary and default vocabulary;
If first similarity is more than predetermined threshold value, the corresponding vocabulary of first similarity is deleted, after obtaining cleaning
Described wait to answer a question.
7. answer lookup method as claimed in claim 5, it is characterised in that described after described pair of cleaning wait to answer a question into
Row completion, the step of obtaining described to be answered a question after completion, include:
Word segmentation processing is carried out to waiting described in after cleaning to answer a question, obtains described waiting corresponding vocabulary of answering a question;
The vocabulary and preset group word rule are contrasted, obtain comparing result;
Wait to answer a question described in after being cleaned according to the comparing result completion, to wait to answer a question described in obtaining after completion.
8. such as claim 1 to 7 any one of them answer lookup method, it is characterised in that described to be judged according to preset rules
Whether found in knowledge base with it is pretreated it is described wait to answer a question corresponding target problem the step of include:
Determine in the knowledge base with it is pretreated it is described wait to answer a question belong to same category of same problems, and calculate institute
State the second similarity for waiting to answer a question between any same problems;
If there is second similarity for being more than default similarity, using the corresponding same problems of maximum second similarity as
Target problem;
If there is no the second similarity more than the default similarity, it is determined that wait to answer a question with described in the knowledge base
Corresponding relevant issues;
The third phase for waiting to answer a question between any relevant issues described in calculating is like degree;
If exist more than the default similarity third phase like spend, using maximum third phase seemingly spend corresponding relevant issues as
Target problem;
If there is no the third phase more than the default similarity like spending, default knowledge mapping is obtained;
Based on the knowledge mapping, the problem of traveling through in the knowledge base of answering a question is waited by described;
If obtaining traversing result, the problem of traversing result is corresponded to, is as the target problem.
9. a kind of customer service robot, it is characterised in that the customer service robot includes memory, processor and is stored in described deposit
On reservoir and the answer search program that can run on the processor, when the answer search program is performed by the processor
Realize such as the step of answer lookup method described in any item of the claim 1 to 8.
10. a kind of computer-readable recording medium, it is characterised in that be stored with answer on the computer-readable recording medium and look into
Program is looked for, realizes that answer described in any item of the claim 1 to 8 such as is searched when the answer search program is executed by processor
The step of method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711363315.2A CN108021691B (en) | 2017-12-18 | 2017-12-18 | Answer searching method, customer service robot and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711363315.2A CN108021691B (en) | 2017-12-18 | 2017-12-18 | Answer searching method, customer service robot and computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108021691A true CN108021691A (en) | 2018-05-11 |
CN108021691B CN108021691B (en) | 2021-09-07 |
Family
ID=62074146
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711363315.2A Active CN108021691B (en) | 2017-12-18 | 2017-12-18 | Answer searching method, customer service robot and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108021691B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109087132A (en) * | 2018-07-18 | 2018-12-25 | 国家电网有限公司 | A kind of the customer problem method for pushing and device of knowledge based map |
CN109388695A (en) * | 2018-09-27 | 2019-02-26 | 深圳前海微众银行股份有限公司 | User's intension recognizing method, equipment and computer readable storage medium |
CN109657227A (en) * | 2018-10-08 | 2019-04-19 | 平安科技(深圳)有限公司 | Contract feasibility determination method, equipment, storage medium and device |
CN109684632A (en) * | 2018-12-12 | 2019-04-26 | 深圳前海达闼云端智能科技有限公司 | It is a kind of nature semantic understanding method, apparatus, calculate equipment |
CN109688046A (en) * | 2018-08-21 | 2019-04-26 | 平安科技(深圳)有限公司 | Issue management method, apparatus, equipment and storage medium |
CN109739962A (en) * | 2018-12-26 | 2019-05-10 | 广州灵聚信息科技有限公司 | A kind of control method and device of Chat mode |
CN109947909A (en) * | 2018-06-19 | 2019-06-28 | 平安科技(深圳)有限公司 | Intelligent customer service answer method, equipment, storage medium and device |
CN110008340A (en) * | 2019-03-27 | 2019-07-12 | 曲阜师范大学 | A kind of multi-source text knowledge indicates, obtains and emerging system |
CN110209768A (en) * | 2019-05-31 | 2019-09-06 | 中国联合网络通信集团有限公司 | The problem of automatic question answering treating method and apparatus |
CN110209781A (en) * | 2018-08-13 | 2019-09-06 | 腾讯科技(深圳)有限公司 | A kind of text handling method, device and relevant device |
CN110334272A (en) * | 2019-05-29 | 2019-10-15 | 平安科技(深圳)有限公司 | The intelligent answer method, apparatus and computer storage medium of knowledge based map |
CN110378789A (en) * | 2019-07-16 | 2019-10-25 | 招商银行股份有限公司 | Solution, system and the computer readable storage medium of disputed transaction |
CN110737759A (en) * | 2019-09-06 | 2020-01-31 | 中国平安人寿保险股份有限公司 | Evaluation method and device for customer service robot, computer equipment and storage medium |
CN110750625A (en) * | 2018-07-04 | 2020-02-04 | 北京国双科技有限公司 | Judicial question-answering method and related equipment |
CN113157856A (en) * | 2021-03-12 | 2021-07-23 | 北京英大融实科技有限公司 | Customer service knowledge base data processing method and device and storage medium |
CN113704434A (en) * | 2021-09-01 | 2021-11-26 | 内蒙古大学 | Knowledge base question and answer method, electronic equipment and readable storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103425640A (en) * | 2012-05-14 | 2013-12-04 | 华为技术有限公司 | Multimedia questioning-answering system and method |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104657346A (en) * | 2015-01-15 | 2015-05-27 | 深圳市前海安测信息技术有限公司 | Question matching system and question matching system in intelligent interaction system |
CN105589844A (en) * | 2015-12-18 | 2016-05-18 | 北京中科汇联科技股份有限公司 | Missing semantic supplementing method for multi-round question-answering system |
CN106484801A (en) * | 2016-09-23 | 2017-03-08 | 厦门快商通科技股份有限公司 | A kind of dialogue method of intelligent customer service robot and its knowledge base management system |
CN106663130A (en) * | 2016-06-29 | 2017-05-10 | 深圳狗尾草智能科技有限公司 | A question and answer method, system and robot |
KR101769610B1 (en) * | 2016-02-11 | 2017-08-21 | 한국과학기술원 | System and method for query graph construction based on question-type |
-
2017
- 2017-12-18 CN CN201711363315.2A patent/CN108021691B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103425640A (en) * | 2012-05-14 | 2013-12-04 | 华为技术有限公司 | Multimedia questioning-answering system and method |
CN104050256A (en) * | 2014-06-13 | 2014-09-17 | 西安蒜泥电子科技有限责任公司 | Initiative study-based questioning and answering method and questioning and answering system adopting initiative study-based questioning and answering method |
CN104657346A (en) * | 2015-01-15 | 2015-05-27 | 深圳市前海安测信息技术有限公司 | Question matching system and question matching system in intelligent interaction system |
CN105589844A (en) * | 2015-12-18 | 2016-05-18 | 北京中科汇联科技股份有限公司 | Missing semantic supplementing method for multi-round question-answering system |
KR101769610B1 (en) * | 2016-02-11 | 2017-08-21 | 한국과학기술원 | System and method for query graph construction based on question-type |
CN106663130A (en) * | 2016-06-29 | 2017-05-10 | 深圳狗尾草智能科技有限公司 | A question and answer method, system and robot |
CN106484801A (en) * | 2016-09-23 | 2017-03-08 | 厦门快商通科技股份有限公司 | A kind of dialogue method of intelligent customer service robot and its knowledge base management system |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109947909A (en) * | 2018-06-19 | 2019-06-28 | 平安科技(深圳)有限公司 | Intelligent customer service answer method, equipment, storage medium and device |
CN109947909B (en) * | 2018-06-19 | 2024-03-12 | 平安科技(深圳)有限公司 | Intelligent customer service response method, equipment, storage medium and device |
CN110750625B (en) * | 2018-07-04 | 2022-05-27 | 北京国双科技有限公司 | Judicial question-answering method and related equipment |
CN110750625A (en) * | 2018-07-04 | 2020-02-04 | 北京国双科技有限公司 | Judicial question-answering method and related equipment |
CN109087132A (en) * | 2018-07-18 | 2018-12-25 | 国家电网有限公司 | A kind of the customer problem method for pushing and device of knowledge based map |
CN110209781A (en) * | 2018-08-13 | 2019-09-06 | 腾讯科技(深圳)有限公司 | A kind of text handling method, device and relevant device |
CN110209781B (en) * | 2018-08-13 | 2023-04-07 | 腾讯科技(深圳)有限公司 | Text processing method and device and related equipment |
CN109688046A (en) * | 2018-08-21 | 2019-04-26 | 平安科技(深圳)有限公司 | Issue management method, apparatus, equipment and storage medium |
CN109688046B (en) * | 2018-08-21 | 2022-08-19 | 平安科技(深圳)有限公司 | Problem management method, device, equipment and storage medium |
CN109388695B (en) * | 2018-09-27 | 2022-11-25 | 深圳前海微众银行股份有限公司 | User intention recognition method, apparatus and computer-readable storage medium |
CN109388695A (en) * | 2018-09-27 | 2019-02-26 | 深圳前海微众银行股份有限公司 | User's intension recognizing method, equipment and computer readable storage medium |
CN109657227A (en) * | 2018-10-08 | 2019-04-19 | 平安科技(深圳)有限公司 | Contract feasibility determination method, equipment, storage medium and device |
CN109684632A (en) * | 2018-12-12 | 2019-04-26 | 深圳前海达闼云端智能科技有限公司 | It is a kind of nature semantic understanding method, apparatus, calculate equipment |
CN109739962A (en) * | 2018-12-26 | 2019-05-10 | 广州灵聚信息科技有限公司 | A kind of control method and device of Chat mode |
CN110008340A (en) * | 2019-03-27 | 2019-07-12 | 曲阜师范大学 | A kind of multi-source text knowledge indicates, obtains and emerging system |
CN110334272B (en) * | 2019-05-29 | 2022-04-12 | 平安科技(深圳)有限公司 | Intelligent question-answering method and device based on knowledge graph and computer storage medium |
WO2020237856A1 (en) * | 2019-05-29 | 2020-12-03 | 平安科技(深圳)有限公司 | Smart question and answer method and apparatus based on knowledge graph, and computer storage medium |
CN110334272A (en) * | 2019-05-29 | 2019-10-15 | 平安科技(深圳)有限公司 | The intelligent answer method, apparatus and computer storage medium of knowledge based map |
CN110209768B (en) * | 2019-05-31 | 2021-08-10 | 中国联合网络通信集团有限公司 | Question processing method and device for automatic question answering |
CN110209768A (en) * | 2019-05-31 | 2019-09-06 | 中国联合网络通信集团有限公司 | The problem of automatic question answering treating method and apparatus |
CN110378789A (en) * | 2019-07-16 | 2019-10-25 | 招商银行股份有限公司 | Solution, system and the computer readable storage medium of disputed transaction |
CN110737759A (en) * | 2019-09-06 | 2020-01-31 | 中国平安人寿保险股份有限公司 | Evaluation method and device for customer service robot, computer equipment and storage medium |
CN110737759B (en) * | 2019-09-06 | 2023-07-25 | 中国平安人寿保险股份有限公司 | Evaluation method and device of customer service robot, computer equipment and storage medium |
CN113157856A (en) * | 2021-03-12 | 2021-07-23 | 北京英大融实科技有限公司 | Customer service knowledge base data processing method and device and storage medium |
CN113704434A (en) * | 2021-09-01 | 2021-11-26 | 内蒙古大学 | Knowledge base question and answer method, electronic equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108021691B (en) | 2021-09-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108021691A (en) | Answer lookup method, customer service robot and computer-readable recording medium | |
CN105701088B (en) | The method and apparatus manually talked with are switched to from machine dialogue | |
US9082084B2 (en) | Facilitating machine learning in an online social network | |
Shi‐Nash et al. | Data analytics and predictive analytics in the era of big data | |
CN109145101B (en) | Man-machine conversation method, device and computer readable storage medium | |
US20150278813A1 (en) | Determining a temporary transaction limit | |
CN109816483B (en) | Information recommendation method and device and readable storage medium | |
US11887184B2 (en) | Techniques to facilitate online commerce by leveraging user activity | |
CN111538844A (en) | Target field knowledge base generation and problem solution method and device | |
CN106530010A (en) | Collaborative filtering method and apparatus based on fusion of time factor | |
CN109063000A (en) | Question sentence recommended method, customer service system and computer readable storage medium | |
CN112966189B (en) | Fund product recommendation system | |
CN117608650B (en) | Business flow chart generation method, processing device and storage medium | |
CN107833088A (en) | Content providing, device and smart machine | |
US20110208753A1 (en) | Method and Apparatus for Computing Relevancy | |
CN113656558A (en) | Method and device for evaluating association rule based on machine learning | |
CN109493186A (en) | The method and apparatus for determining pushed information | |
CN111222769A (en) | Annual report data quality evaluation method and device, electronic equipment and storage medium | |
CN110717095B (en) | Service item pushing method and device | |
CN116468096B (en) | Model training method, device, equipment and readable storage medium | |
CN112991009A (en) | Object recommendation method and device, electronic equipment and storage medium | |
CN112862008A (en) | Training method of preference prediction model and prediction method of user preference | |
Pla et al. | Context-aware case-based reasoning | |
CN104966212B (en) | The detection method and application process of user's mobile degree | |
CN117217852B (en) | Behavior recognition-based purchase willingness prediction method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |